Prepare and test your skills
Prepare and test your skills
Free sample question, answered and explained below. The Generative AI Leader practice bank has 2,191 more like it. Try 10 free: instant score, no account.
A news media company wants to build an AI assistant that delivers summaries of breaking international news and current market developments. However, the standard foundational large language model frequently hallucinates details or fails to provide answers because current events occurred after its training cutoff date.
Which Vertex AI grounding capability should the company implement to connect the model to real-time public web information and mitigate hallucinations?
Vertex AI Feature Store is a managed repository designed to store, serve, share, and track structured machine learning features for predictive models (such as recommendation engines, tabular models, and fraud detection algorithms).
Parameter-Efficient Fine-Tuning (PEFT), such as Low-Rank Adaptation (LoRA), is a training technique that adapts a pre-trained model to specific downstream tasks or stylistic formats by updating only a small subset of model parameters.
Temperature is an inference hyperparameter that controls the randomness and creativity of a large language model's token sampling distribution during text generation. Lowering the temperature makes the model choose higher-probability words, resulting in more deterministic and focused responses.
Grounding with Google Search is a capability within Google Cloud's Vertex AI platform that links foundation models directly to Google's live web search index. Instead of relying exclusively on static pre-trained parameters, the model dynamically queries the web to retrieve the most up-to-date factual information before generating a response.
For tracking live world events and breaking public news, Grounding with Google Search is the native Google Cloud solution designed specifically to fetch external, real-time internet data and eliminate temporal knowledge gaps without custom search pipeline engineering.
Answer them yourself and get your score the moment you finish. No account needed.
Full access includes
Keep the momentum going with these hand-picked practice scenarios
Last updated